American Finally Commits, DFW’s Expansion Can Begin

American, DFW - Dallas/Ft Worth

There has been discussion about expanding Dallas/Forth Worth International Airport probably since the day it opened. The airport has expanded since that time, but now, a commitment has been made by the airport and American to build Terminal F. Sure, this will add sorely-needed gates as American marches toward 1,000 daily departures. But more importantly, it will create symmetry in the airport’s terminals, giving tremendous relief to those with OCD.

When DFW was built, there was a plan to continue building semi-circular terminals both north and south of the original buildings. You can see this in the way the terminals were originally numbered. Terminal B was 2W, Terminal A was 2E, Terminal C was 3E, and Terminal E was 4E. You would have assumed that someday we would see a 1W, 1E, 3W, and 4W.

So far, all we’ve seen built (other than a few gates tagged on to existing terminals) is Terminal D (what would have been 3W previously) as well as a satellite gate building down near Terminal E. Terminal D was constructed for international operations in 2005, and that was the biggest expansion since the airport opened. The airport has also done extensive renovations on Terminals A, E, and B. Terminal C, however, remains a relic of another time.

While renovations are great, American has continued to grow and the airport is squeezed for gates. American already controls all of A, B, C, and a good chunk of D. It took a few gates in E and just this year it opened a new regional operation in that Terminal E satellite. In other words, part of Terminal D is for other international airlines while Terminal E is pretty much everyone else.

The airport has needed to either renovate or destroy Terminal C and build new gates elsewhere for some time. It’s not that the airport didn’t know this. American just couldn’t commit to a project.

At one point there was discussion about building a whole new terminal complex that was better suited for connections, but that would have been incredibly expensive. There has been talk about building a Terminal F and then knocking down C entirely. But every time these discussions progressed, American wouldn’t commit.

Now as American tries to push DFW to 1,000 daily flights and beyond, it was obvious that something needed to happen far earlier. Only now has American agreed to do something, but this is a very reactive move. This decision should have been made long ago.

So what is going to happen? We don’t know. It looks like American effectively said “ok, we want to keep at our costs at a certain level, so uh, what does that get us? Find us something good, and we’ll sign a new lease.” With that, the airport brought forward options in the $3 to $3.5 billion range, but there is nothing concrete yet.

In that range, American can get a renovation of Terminal C and at least part of a new Terminal F. The Terminal C renovation was supposed to cost at least $1 billion, so the question is, how much of a new terminal can they build for $2 to $2.5 billion? I guess we’ll find out.

This is all pending the airline coming to a long-term lease agreement with the airport which is just a matter of time. In the meantime, the airport is going to start pushing forward on design options so it can begin the long process. It’s funny, because when a project like this is announced, there’s usually some kind of rendering. But this is all we get:

If anyone is trying to glean some kernel of intel from this, don’t bother. Yes the orange box is square like Terminal D, unlike the round curves of the other terminals. That means absolutely nothing. This is simply the airport trying to put together any kind of media for distribution without have any information at all.

So what will it end up looking like? I don’t know. Didn’t we just talk about this? Stop asking. That being said, I would be surprised if the new terminal looked different than the others. American is growing DFW into a massive connecting airport, and the original terminals are not designed to make connections efficient. I have to think that Terminal F will do something to fix that.

Now, we wait. Once a lease is signed and a design is chosen, they can start doing the work. At the earliest, this could be done by 2025. That is six long years for American to operate in an airport where it’s already constrained. This decision should have been made earlier, but I suppose better late than never.

Get Cranky in Your Inbox!

The airline industry moves fast. Sign up and get every Cranky post in your inbox for free.

49 comments on “American Finally Commits, DFW’s Expansion Can Begin

  1. CF, it already takes a month to get to a gate after landing. Another terminal, more flights equals more taxi time. Love Field is packed, DFW is packed. Why wouldn’t they open up Alliance Airport in Fort Worth to some passenger traffic?

    1. Who would move to FWA though? Frontier isn’t big enough to make any difference if they move. The EAS(ish) carriers have to be in the same airport as AA for connections. Spirit would be a great candidate to move (a bunch of flights, focus on costs, and doesn’t care about connections) but they like primary airports, to the point that they’re using AUS’s main terminal when it’d absolutely be cheaper to use the South Terminal.

      Capping Love’s gate count is coming back to haunt the Metroplex.

      1. Capping Love Field gates is not haunting the Metroplex. Plenty of room for growth at DFW. Like Cranky said there is room for even more new terminals.

        Alliance is not an option because there is an agreement that it will only be used for Cargo operations.

    2. Elmer – Well, American and Southwest would have no interest in that. So maybe you could convince another airline to use Alliance, but I wouldn’t expect it. And it wouldn’t really relieve pressure on the other airports.

  2. Well all this is good… but what about services and security… To catch a 5:00 flight t Chicago they must have been more than 100 people in line for the security check: Only one Tsa security officer!!! :-(

    Sent from The #1 Brickoven Maker: Renato

  3. I thought DFW had been designed from the beginning for connecting traffic. Did Dallas really think they’d need an airport that large for local traffic only? It’s certainly not a *modern* hub airport (it’s way short on regional gates), and it needs some improvements, but that’s very different.

    1. Uh, no. The whole terminal concept (circa 1968) was based on parking as close as possible to your gate. There was no thought of connections at all in the original design.

      1. Oh, ok. Learn something new every day. Those city leaders were *ambitious*!

        Certainly if I’m flying from my home airport of RDU to out West, I usually gotta connect, and DFW is my first choice. ORD is generally in poor shape, and is an overcrowded disaster zone during any sort of inclement weather, ATL is out of my way, and while DTW is a nice airport, there are less choices of things to keep me occupied if I’m there for any length of time.

  4. Brett:

    A minor additional point. Just to the north of Terminal A (2E) across the north crossover taxiway there is a small, “Remote North” BizAv Terminal (with exits onto the north crossover taxiway and the north holding area). This area was originally used as auxiliary aircraft parking/gates with bus connections to Terminal A. Corporate aircraft had previously been randomly parked where Terminal F will now be built.

    Also, the elevated train (TrAAin) which connects the terminals was a later addition so for many years, connections between the terminals were very awkward—connecting walkways (added over time), bus and a quirky below grade electric car on rubber wheels.

    And a quick mention of the Terminal E “Satellite” controversy. When it was first proposed and built (DL then had a large DFW presence and wanted more gates and competitive scale versus AA), there was a hue and cry about how it would spoil the “symmetry” of the horseshoe-shaped “Six Terminal” design when viewed from the air. As it turned out, no one seemed to notice. It was and remains “odd” and inconvenient with no direct access to the road system and connection via an awkward tunnel under the ramp.

    I always enjoy your column—keep “ Crankin’ “.

    1. Bob – I didn’t mention that, but yes, that is on what would have been 1E if a full terminal was built, I believe. That was used for remote gates for Eagle at one point, I seem to recall. It wouldn’t surprise me to see another airline like Spirit try to use that at some point.

      1. Brett:

        Yes. Eagle did use the 1E location as well until AA acquired more gates and ramp space in the main terminals. As you say, unless more gates are built, 1E may indeed be used by others like Spirit.

  5. “At one point there was discussion about building a whole new terminal complex that was better suited for connections”

    Really? The complex is already pretty good for connecting in 95% of cases. Get the to nearest train stop, quick train ride and you are there. average gate to gate time must be under 15 minutes unless your walking speed is incredibly slow.

    The bigger problem with DFW is the taxi time – Landing on 17L to the gates is absurdly long, nearly as bad as landing on the Polderbaan at Schipol

    1. Forgot to add the key point: the “bang for buck” of improving the in terminal connection experience must be super super low from rebuilding the terminals- realistically lowering it from 14 minutes to 13 minutes is not gonna improve a great deal, and they could probably manage the same thing by resignalling the train track and adding more vehicles to cut the wait time.

    2. DFW is horrible for connections. Long walks and a horrible train that only goes in one direction. Once AA got rid of More Room Throughout Coach, I vowed to never fly AA again and I haven’t for 15 years.

      I’ll take something with gates on both sides and linear like ATL, DEN, or DTW. Much easier on connections.

        1. Agreed that Skylink makes it actually quite easy, as (unlike DEN and some other airports), all the gates are actually very close to Skylink other than the remote terminals.

          Before Skylink was built was a completely different story.

      1. Arrived in DEN on UAX, at one end of the terminal. Connected to mainline, which “conveniently” departed from the furthest gate st the other end of the terminal. Talk about an inconvenient connection.

      2. I don’t know what you’re talking about. SkyLink moves in both directions and is incredibly convenient for connections – much more so than DEN or DTW.

  6. The real issue is how much DFW can be economically expanded as an airline hub. Let’s not forget that AA said that 3 hubs – CLT, DCA and DFW are their most profitable hubs. Any company would like to add as much capacity to its most profitable hubs – but there are limits to how large and how fast that capacity can be added. AA is adding double digit capacity to many DFW markets this summer even while its fleet plan involves no additional new mainline for years to come. Growth at DFW for AA has to come from capacity cut elsewhere other than from adding seats to existing aircraft.
    DFW is also a costly hub for AA to operate. Parker has said that and it isn’t hard to see why that would be the case given the need to move bags between currently five and potentially six separated terminals with ticket counters spread out across the spread out operation.
    DFW is also significantly south of the largest east-west traffic flows in the US, off of the eastern US where the number of cities is the greatest, and south of the largest US-intercontinental traffic flows. The growth of DFW cannot overcome the network losses in other regions as AA’s least profitable hubs have to close or be significantly reduced – if AA wants to bring its margins up to even industry average levels.

    DFW might grow as a hub for AA but it is an inherently inefficient airport from a labor standpoint, the airport is geographically not in the best position as the primary airport for a global airline, and the growth of AA at DFW is really driven by the lack of profitability in a number of its other hubs.

    1. If DFW is too far south, what does that make MIA? Or IAH? or PHX? If it was so suboptimal, it would be less of a fortress hub than DEN (spoiler: UA’s posiiton in DEN is more tenuous than AA’s at DFW by a country mile). SLC is a fortress hub, sure, but that’s also half the size of AA’s DFW operation.

      As for “too far south to serve cities in the northeast”, AA has JFK/LGA/DCA for local traffic, CLT for connecting (yes it’s mid-atlantic but no worse from that perspective than IAD), and PHL/ORD for a bit of both. Of those, DCA and CLT are more profitable, and traffic doesn’t have to head down south unless it needs to.

      Or are you wanting STL back? :)

      1. MIA serves primarily Latin American and Caribbean connecting traffic. No one flies LAX-MIA-JFK.
        IAH is a lot like DFW and has the same efficiency problems.
        PHX as an AA hub only really exists because of legacy US.

        Calling DFW a fortress is a bit odd. Sure AA has the vast majority of the local traffic share there, but if you include DAL numbers it dilutes significantly, especially with how many fewer WN passengers connect.

    2. No offense, but this sounds like the same logic that insisted hubs at MCI and STL would be the optimal solutions for funneling connecting traffic. How did that work out for the airlines that tried it? Geography is only part of the answer; it’s been shown time and again that you need sufficient O&D to run a hub. Where would you suggest putting the hub instead? It’s not like you can squeeze a bunch more traffic through ORD, and STL’s already been tried and failed. Maybe CLT, but the metro area is a fraction of the size of DFW and I doubt O&D could support a DFW-sized hub.

      Given all that, and the fact that any major airport in the US is ~4 hours block time from DFW, if it’s a “suboptimal” location, it seems like the best option on the table.

    3. Quoting Tim Dunn: “DFW is also significantly south of the largest east-west traffic flows in the US, off of the eastern US where the number of cities is the greatest, and south of the largest US-intercontinental traffic flows.”

      To which I respond: I guess that depends on how one defines the word “significantly.” The linked graphic below does show that DFW is south of the absolute center of the traffic flow across the U.S. but I certainly wouldn’t call the deviation “significant.” The cost to build a completely new connecting airport at the precise geographic center of the country’s air traffic flow would be staggering – and useless. The absolute center of the main east-west flow appears to be northwest of Memphis, which was dehubbed by Delta in favor of Atlanta. The overall migration pattern over the last 50+ years has been toward the south and west, so if that pattern persists, DFW will get closer to the center of the flow, not farther from it.

      Critics can always find a reason to knock people and companies. American has what it has. Its only real choice is to optimize it. That’s what it’s doing.

      1. Nobody says that a hub should be built at the geographic center but that DFW is not in the best location for the largest hub for an airline. Of course MIA and IAH are not in the center of the country but no one is talking about them as 1000 flight/day hubs either.

        And none of you 3 addressed the issue that AA is in a no-growth mode for its mainline fleet which means its DFW growth comes from somewhere else.

        Ultimately, AA’s ability to keep building DFW is linked to its overall financial health.

        I submit that AA’s ability to gain more connecting revenue via DFW is less than the shrinking revenue it is seeing in other hubs including NYC and CHI.

        1. Where is the ideal location then? Spending money on expanding DFW is a much better use of it than expanding a different hub or building a new one.

        2. Atlanta isn’t in the center of the flow but it does pretty well as Delta’s biggest hub. Most people I know don’t mind going a little out of their way to connect (as long as it’s not ridiculous). Houston is near the center of the country if one is going east or west (as is DFW). Pre-merger Continental’s two main hubs (Houston and Newark) were near the periphery of the country, not in the middle. Miami is more of an international gateway than a domestic hub.

          Are Delta and United growing their fleets at a net rate of 10% per year? No. Major airlines aren’t in a growth mode, which is normal for a mature industry. Most major airline growth is coming from upgauging. Based on American’s investor relations guidance, its regional fleet isn’t growing much, from 595 aircraft (which is down from past years) to 614 (with some modest upgauging) over the next three years, which is about the top end of its scope. Mainline is going from 956 to 961 aircraft in the same period.

          Obviously, the growth at DFW has to come from somewhere. But according to a post I saw on, an American executive noted that the only two unprofitable American hubs are LGA (-0.8%) and LAX (-0.9%). JFK’s margin is 3.1%. The next highest is Miami at 6.7%. In fact, four of the top six performing American hubs are not from AMR (as the pre-merger pundits argued) but from US Airways* (PHX* at 9.6%, PHL* at 11.4%, ORD at 11.4%, DFW at 12.7%, DCA* at 13.6%, and CLT* at 14.3%. I’m guessing (and it’s just that) American feels New York is fine, as is probably margin accretive to the network overall, as is LAX, whose losses are something it can probably live with. Again, this is from a post on from a few weeks ago, so one can take the information with as big a grain of salt as one wishes.

          1. DFW is probably AA’s best choice. No one suggested that AA build elsewhere. The point is that DFW does have geographic limitations and that the facility itself was not built from the ground up to be a super connecting hub as ATL was and thus there are a lot of compromises that do work, but come at a cost. ATL is also in the middle of the key eastern US flows and also works far better than any other hub to Europe outside of the NE and also better than any other hub than Miami to Latin America – based on DL’s abililty to sustain its size to those regions from ATL.
            As for the margin of AA hubs, you need only look at where AA is growing and where it is retracting. ORD is geographically ideally located – but AA can’t make international longhaul work there on a year round basis outside of LHR. ORD suffers from AA and UA’s domestic competition which makes neither hub as large as it could be if ORD was a one airline hub. AA continues to shrink in NYC – actually at JFK more than LGA – and regardless of the current runway cutbacks which other carriers are managing to grow through.

            CLT Is similarly located geographically but is a much smaller local airport and the facility is not near as ideal as a hub as ATL or even ORD or DEN. UA realizes that IAH is not as geographically ideal which is why it is not growing as fast as DEN and is focused on international markets which support the local Houston economy more than trying to be a super hub.

            Margin accretive and margin positive on a standalone basis are two very different things. AA is making a lot of decisions about its network now based on the value of the future of those flights to its network; they did with that Asia, losing hundreds of millions of dollars flying to Asia and now have come to the realization that DL and UA with acquired presences in Asia have done a far better job of building their own networks. DL happens to make money flying to Asia while UA does not – according to DOT data.

            AA’s network to Europe is heavily summer seasonal, including many of its flights from DFW.

            AA’s growth at DFW is their best choice from the hubs they have but that doesn’t mean it is the best an airline including AA could do if they built from the ground up. It also means that AA’s size at DFW is tied to its hub performance in the rest of the US and ultimately AA’s financial strength. There is no assurance that AA’s growth at DFW will be sustainable.

            1. That’s interesting. Your original comment essentially argued that American shouldn’t build up DFW. Now you’re changing your tune. Apparently, the airline (whose executives have far more data at their disposal than we do) thinks expanding DFW is economically beneficial or else it wouldn’t be doing it. Of course, there’s no assurance that DFW’s growth will be sustainable. Nothing is guaranteed in life or business. The idea that an airline has to be everything to everybody has died – along with many defunct airlines that used to think that way.

          2. isn’t it ridiculously sad that AA has Hollywood in their back pocket and can’t make LAX even positive returns THERE ?

            and LGA even worse than JFK ? yikes maybe all those 50-seater ERJ slot squatting on super short -hauls aren’t exactly paying off, from either LGA or JFK.

            the number are generally reasonably, although you’d think DFW should have a sizable gap over ORD, cuz their approach of ORD over the last 5 years at the very least suggests anything but yields remotely similar to PHL/DFW.

            Let’s ignore their JFK for a second, but I can’t fathom how their LGA is THAT bad.

            1. I’m guessing that a lot of American’s losses in L.A. are probably related to its Pacific operation. LAX is also very fragmented. I tend to doubt that any carrier is swimming in profits there (except maybe Southwest), but it’s part of being relevant. If the above figures are accurate, both LAX and LGA have losses of less than 1%. Neither are among American’s larger hubs, so I’m guessing the airline feels the relatively small losses there can be subsidized by other flying. If O’Hare’s numbers are accurate, the Chinese flying must have been REALLY unprofitable. It should be quite interesting to see how all of this shakes out.

  7. One obvious way to reduce inter-terminal travel time at DFW would be to speed up the Skylink trains…modify or replace them so they travel faster.

  8. I was just curious if anyone else couldn’t see the second image after “But this is all we get:”. It would’ve been after the seventh paragraph following the overhead photo with the existing layout and brief history.

    I tried on my phone (Android using Chrome), and my computer (Mac using Chrome, Firefox, and Safari).


  9. Mark my words: In or around 2025 we will be hearing about McKinney National opening for commercial service and Southwest will be one of if not the first tenant. DFW will be maxed out with F, Love is maxed out by Federal law, there is no where else to go. TxDOT is building out bypasses and loops around 380 and I75, and low and behold, one of those new freeways will be a connection directly to TKI. The Love Field 5 party agreement expires in 2025, DFW finishes expansion in 2025, the freeway will be done by 2025.

    If LA can support 5 commercial airports, DFW can support 3.

  10. CF – Great article as always!

    All – Great discussion as always!

    My add – Seems US3 are all working to create Fortress Hubs in order to maximize profitability as they limit fleet growth to low single-digits. This leaves Airport Owner-Operators with only a single tenant, albeit at least a profitable one. Little wonder that in this situation (eg DFW), the Airport Owner-Operators require major commitments (financing, duration, terms) from their one customer. Economics 101, right?

    When the Airport-Owners are Local Governments, then I STRONGLY question the benefits for their voters/consumers of having Fortress Hubs and their resulting higher fares. LAX has massive O&D and is the opposite of a Fortress Hub with no airline having 20% share. The result is some of the lowest fares in the country, which benefits the consumer at the expense of the airline.

    And to provoke CF to reply, I believe LGB is forced to price in LAX’s shadow, thus it doesn’t matter whether it is a Fortress Hub (mostly JetBlue) for its 50 flights/day or instead a Mixed Hub (Southwest, JetBlue, others), right? I don’t follow LGB City Council, but was this part of their rational for not supporting JetBlue’s request for International Flights/Facilities, or was it all driven by NIMBY/Noise?

    Self-Disclosure: I am an airline customer, not an Airline Revenue/Yield Management Employee or a Financial Analyst who covers the Airlines. Would be great if others would describe their role that drives their point-of-view given the old adages that “One’s perception is based on one’s position” and “Facts all come with a point of view”

    1. LAX – But DFW really isn’t a fortress hub. You have Spirit with a sizable operation that brings low fares to major cities. Then you also have Southwest at Love Field which at least tempers fares to some extent. DFW is a great place to live for air service.

      As for Long Beach, it was all about the NIMBYs. But it’s true that LGB doesn’t control its own pricing. It is wedged in between other airports that really set things.

      1. “But DFW really isn’t a fortress hub. You have Spirit with a sizable operation that brings low fares to major cities. Then you also have Southwest at Love Field which at least tempers fares to some extent. DFW is a great place to live for air service.”

        i really hope you’re either just obviously joking, or at least got well compensated as a spokes person for your world class acumen in media spin.

        If you seriously think the 4.74% market share Spirit has at DFW or the 140-160-ish daily departure focus ops that is DAL for WN has ANY impact against the 900+ behemoth that is AA DFW where they have minimum 80% market share (, allocate Mesa whatever you like) and a good lead over 2nd largest hub by departures (

        On top of DAL gates cap , the mid-con nature of Dallas means regardless of tech upgrades at WN, it’s not a place where you can milk tons of additional turns per gate via red-eye ops. 18 gates with world-class turns of 8.5 per daytime has a realistic cap of ~150-ish for WN, give or take.

        But if you actually believe that statement you’ve made, there are some folks at FRA, YYZ, DTW more than willing to see you their used cars. Between your earlier stringent defense of 737max and now this, i’m just guessing you stopped bothering to distance yourself from any current (or lingering) affinity to both AA and WN.

        1. henry LAX – I’ll ignore the sniping and focus on the facts.

          I don’t have operations by carrier at my fingertips, so I’ll stick with passengers. Looking at both Love and DFW for March, American has 69 percent of the passengers while Southwest has 18 percent. Everyone else is much smaller, obviously. Is American easily the dominant hub carrier?
          Absolutely. But I wouldn’t call it a fortress hub in the way I’d call Charlotte a fortress hub…. or the way I would have called DFW a fortress hub 15 years ago.

          From a passenger perspective, there is competition on nearly all of the most important routes and fares have fallen significantly. It’s been pronounced enough to get the local media to even notice. I was quoted in this article about it back in February.

          If there’s an official definition of the term “fortress hub,” I’m unaware of it. So this could be entirely semantics. You may define it differently than I do. But in my mind, this isn’t the fortress hub it used to be.

          1. CF is correct that AA has a less dominant position of the Metroplex’ two airports than other carriers have in other cities.

            According to DOT data, of the LOCAL MARKET, not those passengers that are connecting in N. Texas, AA has a 52.2% local market share compared to 23.6% for WN, 6.9% for DL and 6.5% for NK.

            If you include only N. America passengers, AA falls to 51.8 while WN is at 24.8.

            DL at ATL – with a shared “hub” with WN – is 8 points higher and a larger number of passengers and revenue while at CLT – which is probably the best example of a fortress hub, AA”s N. America share is over 70% – but the local market is half the size of AA at DFW.

            These local market numbers show that AA’s capacity additions means it is growing capacity far faster than the local market which means it either has to sell nearly all of the new capacity as connections or it will push fares in the local market down because the growth rate is far above the growth rate for the local market.

  11. I thought connections were pretty convenient with the train, but I don’t have reason to go through there very often.

  12. i’ll volunteer to do Theresa May’s thankless job any summer day in hell before i’ll subject myself to yet another one of those later afternoon t-storms that DFW is famous for.

  13. Perhaps we just need some running tracks in the terminals. They are certainly less costly than a new terminal and will be immensely helpful next time AA sells me a ticket with 35 min DFW connection. (35 min connection) minus (20 min taxi) minus (10 min cabin door close) = 5 min. Yeah it’ll totally work!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Cranky Flier